Measures of physical activity using cell phones: validation using criterion methods.

摘要: Background: Physical activity is associated with reduced risks of many chronic diseases. Data collected on physical in large epidemiological studies often based paper questionnaires. The validity these questionnaires debated, and more effective methods are needed. Objective: This study evaluates repeated measures level (PAL) the feasibility using a Java-based questionnaire downloaded onto cell phones for collection such data. data obtained were compared reference estimates doubly labeled water method indirect calorimetry (PALref). Method: Using phone application, 22 women reported their two short questions answered daily over 14-day period (PALcell). Results from also against levels assessed by regular completed at end (PALquest1 PALquest2). PALcell, PALquest1, PALquest2 PALref Bland Altman procedure. Results: mean difference between PALcell was small (0.014) narrow limits agreement (2SD = 0.30). Compared PALref, PALquest1 (0.004 0.07, respectively); however, wider (PALquest1, 2SD 0.50 PALquest2, 0.90). test trend statistically significant (slope regression line 0.79, P .04) as well 1.58, < .001) when PALref. Conclusion: A administered produced PAL that agreed values. Furthermore, phones, values narrower than corresponding may be feasible cost-effective large-scale prospective activity.